3448

J. Agric. Food Chem. 2003, 51, 34483454

Distribution of Hydroxycinnamic Acid Conjugates in Fruit of

Commercial Eggplant (Solanum melongena L.) Cultivars
BRUCE D. WHITAKER*,

AND JOHN

R. STOMMEL

Produce Quality and Safety Laboratory and Vegetable Laboratory, Beltsville Agricultural Research
Center, Agricultural Research Service, USDA, 10300 Baltimore Avenue,
Beltsville, Maryland 20705-2350

There is gathering evidence that antioxidant phytonutrients in fruits and vegetables have healthpromoting effects. Eggplant fruit have a high content of antioxidant phenolic compounds. We evaluated
the main class of eggplant phenolics, hydroxycinnamic acid conjugates, in the fruit of seven commercial
cultivars. Fourteen conjugates were quantified and identified by high-performance liquid chromatography, ES--MS, and 1H NMR data. Significant differences in their content and composition were
evident among cultivars and in tissue from stem, middle, and blossom end segments. Chlorogenic
acid (5-O-caffeoylquinic acid) was the predominant compound, and its 3-O-, 4-O-, and 5-O-cis isomers
were also present. The 10 other phenolics fell into four groups, including 3,5- and 4,5-dicaffeoylquinic
acid isomers, four amide conjugates, two unknown caffeic acid conjugates, and 3-O-acetyl esters of
5-O- and 4-O-caffeoylquinic acid. Dicaffeoylquinic and 3-O-acetyl chlorogenic acids were most variable
among the cultivars. Dicaffeoyquinic acids were most abundant in the blossom end, whereas 3-Oacetyl esters were highest in the midsection.
KEYWORDS: Eggplant fruit; Solanum melongena; chlorogenic acid; caffeoylquinic acid esters; caffeoylpolyamine amides; hydroxycinnamic acid; 3-O-acetyl-5-O-caffeoylquinic acid

INTRODUCTION

The USDA recommends daily consumption of five servings

of fruits or vegetables partly on the basis of accumulated
evidence that the phytonutrients in these horticultural products
are beneficial to human health (1-3). The combined antioxidant
activity of chemical constituents in fruit and vegetable tissues
is thought to be one key factor (4-6). Eggplant is among the
top 10 vegetables in oxygen radical absorbance capacity, and
this is attributed to its high content of phenolic antioxidants
(7). Phenolic compounds extracted from eggplant fruit and
administered orally to normal- and cholesterol-fed rats had a
significant hypolipidemic action (8). Winter and Herrmann (9)
determined that quinic acid esters of hydroxycinnamic acids are
the major class of polyphenols in eggplant fruit, with chlorogenic
acid as the predominant compound. However, this study
included a sampling of only two cultivars and focused exclusively on monohydroxycinnamic acid quinate esters and glucosides. Accurate quantitation, as well as knowledge of the
complete profile of phenolic phytochemicals and their supposed
health benefits, is needed to establish future dietary guidelines
for recommending phenolic-rich foods such as eggplant as
modulators of disease.
* To whom correspondence should be addressed. Tel: 301-504-6984.
Fax: 301-504-5107. E-mail: whitakeb@ba.ars.usda.gov.
Produce Quality and Safety Laboratory.
Vegetable Laboratory.

The cultivated eggplant, Solanum melongena L. (Solanaceae),

is a species of considerable economic importance in many
tropical and subtropical parts of the world. Consumption of
eggplant is growing in the United States as ethnic diversity of
urban areas increases and consumer health and diet-related
education expands. Eggplant is of tropical origin and produces
fruit with many different shapes, sizes, and colors, depending
on the cultivar. Eggplant fruit for human consumption are
harvested when they are physiologically unripe, similar to
cucumbers and winter squash. The purple eggplant is widely
used worldwide, but other varieties that differ in color, size,
and shape are cultivated and consumed. Scant information is
available about the fruit chemical composition and nutritive
value among market varieties of eggplant. Studies including data
on phenolic phytonutrients in the fruit have focused mainly on
changes in the level of total polyphenols during fruit development and postharvest storage or on the role of these compounds
as substrates for polyphenol oxidase in the enzymatic browning
of cut or injured tissue (10-12). In this study, we have examined
the content and distribution of hydroxycinnamic acid derivatives
in fruit flesh of seven commercial eggplant cultivars representing
a variety of market types.
Hydroxycinnamic acids are phenolic acids included in the
large class of secondary plant metabolites known as phenylpropanoids, which are produced via a pathway initiated primarily
by conversion of the amino acid phenylalanine to cinnamic acid
by the enzyme phenylalanine ammonia lyase (13). Typically,

10.1021/jf026250b This article not subject to U.S. Copyright. Published 2003 by the American Chemical Society
Published on Web 04/29/2003

Hydroxycinnamic Acid Conjugates in Eggplant Fruit

hydroxycinnamic acids in fruit tissues are almost entirely
esterified to other polyhydroxylated compounds such as quinic
acid, tartaric acid, and glucose (14, 15). Esters of caffeic acid
predominate in solanaceous species (16), and the most abundant
of these compounds is almost invariably chlorogenic acid (5O-caffeoylquinic acid). Quinic acid can also be esterified with
ferulic acid, p-coumaric acid, and (rarely) sinapic acid, and an
array of isomers as well as di- and trihydroxycinnamoylquinic
acid esters occurs at lower concentrations. Chlorogenic acid and
other caffeoyl esters are among the most potent free radical
scavengers found in plant tissues (17, 18), and chlorogenic acid
acts as an antioxidant agent both in human erythrocytes (19)
and for human low-density lipoproteins in vitro (20). A variety
of caffeoyl esters, most notably diesters of quinic and tartaric
acid, have been shown to have antiviral activity, including
efficacy against human immunodeficiency virus (21, 22). This
class of hydroxycinnamic acid derivatives has also been reported
to possess antimutagenic activity (23) and antimicrobial activity
against several species of human pathogenic bacteria (24). Our
quantitative analysis of hydroxycinnamic acid conjugates in
eggplant fruit included four isomers of monocaffeoylquinic acid,
two isomers of dicaffeoylquinic acid, four partially characterized
amide derivatives, two unidentified caffeic acid conjugates, and
two compounds tentatively identified as novel 3-O-acetyl esters
of caffeoylquinic acid isomers.
MATERIALS AND METHODS
Plant Material and Cultural Methods. Seeds of seven commercial
eggplant (S. melongena L.) cultivars, Black Magic, Classic, Epic,
Ghostbuster, Orient Express, Elondo, and Pirouette, were selected to
represent various eggplant market types. Black Magic, Classic, and
Epic produce large, black/purple-pigmented, globular to tapered fruit
characteristic of the predominant market class in the United States.
Elondo is marketed as a large, cylindrical Italian type with dark purple
elongated teardrop-shaped fruit. Ghostbuster produces unique whitepigmented, mid-sized, elongated, egg-shaped fruit. Orient Express is
an early fruiting cultivar, producing black-pigmented, very elongated,
Asian style fruit. Pirouette is a novel, very early producing cultivar
bearing small teardrop-shaped, purple-pigmented fruit.
After germination of the seed, plants were grown from seedlings in
a greenhouse using standard production practices. Six 7 week old plants
of each accession were transplanted to field plots at the Beltsville
Agricultural Research Center, Beltsville, MD, into Keyport fine loam
soil using standard horticultural practices for eggplant production in
Maryland (25). Plants were spaced at 0.45 m intervals in single rows
on polyethylene-covered raised beds, with beds positioned on 1.5 m
centers. Fertilizer and supplemental water was supplied using trickle
irrigation.
Three fruit from each of three plants for each cultivar were harvested
at market maturity (assessed visually), which ranged from about 3040 days postanthesis. At harvest, the fruit were washed and peeled,
and a 2 cm wide longitudinal section (from stem to blossom end) was
excised from the middle. This tissue slice was divided equally into
upper (stem end), middle (central mesocarp), and bottom (blossom end)
segments, each of which was quickly diced, frozen in liquid N2, and
lyophilized. The freeze-dried tissue samples were pulverized and stored
individually in small Ziploc bags at -80 C.
Tissue Extraction and Sample Preparation. Phenolic acids were
extracted from 0.2 g samples of the lyophilized, powdered tissue by
sonicating for 15 min in 10 mL of methanol containing 0.5% butylated
hydroxytoluene (BHT). The first methanol extract was decanted after
centrifugation, and the tissue sample was extracted a second time with
10 mL of methanol plus BHT. The two extracts were combined, filtered
through Whatman No. 4 filter paper, and then passed through a
Whatman PTFE syringe filter (0.2 m pore size). Filtered extracts were
stored in capped brown glass vials at -80 C until removed for highperformance liquid chromatography (HPLC) analysis.

J. Agric. Food Chem., Vol. 51, No. 11, 2003

3449

A 1.5 mL aliquot of each extract was transferred to a 2 mL amber

HPLC vial, followed by addition of 25 g of sesamol (3,4-methylenedioxyphenol) in 25 L of methanol as an internal standard (26). The
solvent was removed by N2 evaporation at 35 C, and the residue was
resuspended in 1.0 mL of 0.02% (2 mM) phosphoric acid in methanolwater, 1:1 (v/v), with vortexing for 30 s. After centrifugation for 3
min to pellet the insoluble BHT, the supernatant was transferred to a
2 mL microfuge tube and centrifuged at 16 000g for 1.5 min to pellet
remaining particulates. The supernatant was transferred to a new 2 mL
amber HPLC vial, which was flushed with N2 and sealed with a Teflonlined septum cap. When not analyzed immediately, samples were stored
overnight at -80 C.
HPLC Analysis. Phenolics in 20 L injections of the eggplant fruit
extracts were separated and quantified by RP-HPLC using an HP 1100
Series instrument with a quaternary pump, autosampler, and photodiode
array detector (Agilent Technologies). Data were acquired and analyzed
with HP ChemStation software on a pentium PC. A method was
developed to achieve separation of the major constituents on a 250
mm 4.6 mm i.d., 5 m Luna C18(2) analytical column (Phenomenex,
Torrence, CA) within 30 min. A binary mobile phase gradient of
methanol in 0.01% aqueous phosphoric acid was used as follows: 0-15
min, linear increase from 5 to 25% methanol, 1.0 mL/min; 15-25 min,
linear increase from 25 to 50% methanol, 1.0 mL/min; 25-28 min,
50% methanol, 1.0 mL/min; 28-30 min, linear increase from 50 to
100% methanol and from 1 to 1.2 mL/min; 30-32 min, 100% methanol,
1.2 mL/min; 32-36 min, linear decrease from 100 to 5% methanol,
1.2 mL/min; 36-38 min, 5% methanol, linear decrease from 1.2 to
1.0 mL/min. Quantification was based on absorbance at 325 nm relative
to the sesamol internal standard and an external standard of authentic
chlorogenic acid purchased from Aldrich (Milwaukee, WI). An
isochlorogenic acid standard including 3,4-, 3,5-, and 4,5-dicaffeoylquinic acid isomers was obtained from ICN-K&K Laboratories (Plainview,
NY).
LC-MS Analysis. Atmospheric pressure ionization mass spectrometry analysis was performed on a Quattro LC benchtop triple quadrupole
mass spectrometer (Micromass Ltd., Manchester, U.K.) operated using
the electrospray ionization interface in the negative mode (ES-). Mass
spectrometric data were acquired in the full scan mode over the m/z
150-600 range. Sensitivity of the mass spectrometer was optimized
using the chlorogenic acid standard from Aldrich. A Waters 2690 HPLC
system was utilized for separation of phenolics in the eggplant extracts
and caffeoylquinic acid standards. The Luna C18(2) column and mobile
phase gradient were identical to those used for RP-HPLC-UV analysis
with the following exceptions: 0.05% aqueous formic acid was
substituted for 0.01% phosphoric acid, and the initial linear increase
from 5 to 25% methanol at 1.0 mL/min extended from 0 to 12 min
rather than 0-15 min. Eggplant phenolic samples were in methanolwater, 1:1 (v/v), including 0.1% formic acid, and 20 L was injected
per run by a Waters autosampler. Mass spectrometric analyses were
performed with MassLynx 3.5 software (Micromass Ltd). UV spectroscopic data were acquired using a Waters model 996 photodiode
array detector over the range of 210-400 nm.
NMR Spectroscopy. Caffeoylquinic acid standards and major
eggplant fruit phenolics isolated by HPLC were dissolved in 0.8 mL
of CD3OD, and 1H NMR spectra were acquired deuterium-locked at
25 C using a Bruker QE 300 MHz NMR spectrometer. Chemical shift
values were assigned relative to the frequencies of residual nondeuterated water and methanol externally referenced to tetramethylsilane.
Statistical Analyses. The SAS System (SAS Institute, Cary, NC)
was used to perform all statistical analyses. Analysis of variance was
obtained with the SAS General Linear Models procedure with cultivars
treated as fixed effects.
RESULTS AND DISCUSSION

Fourteen compounds that were present in many but not all

fruit and fruit tissue segments were separated by HPLC (Figure
1) and identified or tentatively identified as hydroxycinnamic
acid derivatives (Table 1). Identification was based on HPLC
elution times, UV absorbance spectra, ES--MS mass spectrometric data, and in some cases 1H NMR data. As indicated in

3450

J. Agric. Food Chem., Vol. 51, No. 11, 2003

Whitaker and Stommel

Figure 1. HPLC chromatogram showing separation of the 14 hydroxycinnamic acid conjugates in eggplant fruit extracts that were characterized and

quantified (labeled P-1P-14). Chlorogenic acid (P-6) was invariably the most abundant phenolic acid derivative. The internal standard peak (sesamol)
is highlighted in black. Note that in this chromatogram the two compounds tentatively identified as 3-O-acetyl esters of 5-O- and 4-O-caffeoylquinic acid
(P-10 and P-12, respectively) are at levels 10200-fold higher than those in fruit tissue of the seven cultivars examined.
Table 1. Grouping Based on Identification or Tentative Identification of

the 14 Hydroxycinnamic Acid Conjugates in Eggplant Fruit Extracts

that Were Quantified by HPLCUV and Subjected to Statistical
Analyses
ES--MS
[M 1]ion

elution
time
(min)

UV absorbance
maxima
(280330 nm)

peak 2
peak 6
peak 8
peak 9

17.4
21.8
23.1
24.2

324, 297 (sh)a

326, 296 (sh)
327, 297 (sh)
318

group 1
353
353
353
353

3-caffeoylquinic acid
5-caffeoylquinic acid
4-caffeoylquinic acid
5-cis-caffeoylquinic acid

peak 13
peak 14

28.4
28.7

328, 298 (sh)

328, 296 (sh)

group 2
515
515

3,5-dicaffeoylquinic acid
4,5-dicaffeoylquinic acid

peak 1
peak 3
peak 4
peak 5

10.1
18.8
19.6
20.6

317, 292
320, 289
319, 288
319, 293

group 3
249
470b
470b
468b

N-caffeoylputrescine
dihydroxycinnamoyl amide
dihydroxycinnamoyl amide
N,N-dicaffeoylspermidine

peak 7
peak 11

22.3
25.6

329, 298 (sh)

326, 296 (sh)

group 4
355
ndc

caffeic acid conjugate

caffeic acid conjugate

peak 10
peak 12

24.7
26.1

327, 297 (sh)

325, 296 (sh)

group 5
395
395

3-acetyl-5-caffeoylquinic acid
3-acetyl-4-caffeoylquinic acid

conjugate
identification

a sh, secondary absorbance maximum appearing as a shoulder on the primary

maximum. b [M 1]- ion with even mass number suggests that the structure
includes an odd number of nitrogen atoms. c [M 1]- ion was not determined.

Table 1, these 14 phenolics were grouped into five classes as

follows: chlorogenic acid isomers (group 1), isochlorogenic acid
isomers (group 2), hydroxycinnamic acid amide conjugates
(group 3), unknown caffeic acid conjugates (group 4), and
acetylated chlorogenic acid isomers (group 5).
Group 1 included four compounds with HPLC elution times
of 17.4, 21.8, 23.1, and 24.2 min (Figure 1; P-2, P-6, P-8,
and P-9). They were identified as the 3-O-trans, 5-O-trans, 4-Otrans, and 5-O-cis isomers of caffeoylquinic acid (Figure 2;
neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid,
and cis-chlorogenic acid, respectively). Criteria used for identification were as follows: ES--MS spectra of all four phenolics
had a prominent molecular ion minus a proton, [M - 1]-, at
m/z 353, required for caffeoylquinic acid (C16H18O9 ) 354).
For the 17.4 and 23.1 min caffeoylquinic acids, m/z 353 was
the major ion (100% relative intensity), whereas for the 21.8

and 24.2 min isomers, the quinic acid ion, m/z 191, was
predominant. The HPLC peaks at 17.4, 21.8, and 23.1 min had
closely similar UV spectra, whereas that of the 24.2 min peak
was distinctly different (Table 1). The HPLC elution time, ES-MS spectrum, and 1H NMR spectrum of the highly abundant
phenolic at 21.8 min were identical to those of the chlorogenic
acid (5-O-caffeoylquinic acid) standard. The elution order of
the 17.4 and 23.1 min isomers indicated that they were 3-Oand 4-O-caffeoylquinic acid, respectively (18, 27), and comparison of their 1H NMR spectra with published values
confirmed this (18, 28). Compared with chlorogenic acid, the
24.2 min isomer had a similar ES--MS spectrum but a very
different UV spectrum, which suggested that it was 5-O-ciscaffeoylquinic acid. This was confirmed by the 1H NMR signal
for the olefinic proton on C8, a doublet with J ) 13 Hz at 5.78
ppm, indicative of a cis double bond between C7 and C8 (29).
Group 2 consisted of two phenolics with elution times of
28.4 and 28.7 min (Figure 1; P-13 and P-14), both with UV
spectra very similar to that of chlorogenic acid (Table 1), and
[M - 1]- at m/z 515, required for dicaffeoylquinic acid
(isochlorogenic acid; C25H24O12 ) 516). Elution times and mass
spectra of the 28.4 and 28.7 min isomers were identical to those
of 3,5-di-O- and 4,5-di-O-caffeoylquinic acid standards (Figure
2), which were identified by comparison of their 1H NMR
spectra with published values (30, 31).
Group 3 included four phenolics with elution times of 10.1,
18.8, 19.6, and 20.6 min (Figure 1; P-1, P-3, P-4, and P-5).
Tentative identification of these constituents as hydroxycinnamic
acid amide conjugates was based on their UV spectra, molecular
masses, and 1H NMR data. UV spectra of the 10.1 and 20.6
min peaks were very similar, as were those of the 18.8 and
19.6 min peaks (Table 1). ES--MS of all four compounds in
group 3 gave the [M - 1]- ion at 100% relative intensity (Table
1). This was the sole ion for the 10.1 min phenolic, whereas
mass spectra of the 18.8, 19.6, and 20.6 min compounds
included a sodium adduct ion at 10% relative intensity (m/z
492, 492, and 490, respectively). The almost identical UV and
mass spectra of the 18.8 and 19.6 min compounds indicate that
they are isomers of the same phenolic, and the 20.6 min
compound, with a molecular mass two protons less than these,
may be a related phenolic with an additional double bond.
Furthermore, the even-numbered mass of the [M - 1]- ion of
all three suggests that their structures include an odd number
of nitrogen atoms. Amide conjugates of hydroxycinnamic acids
with tyramine and polyamines are common in solanaceous

Hydroxycinnamic Acid Conjugates in Eggplant Fruit

J. Agric. Food Chem., Vol. 51, No. 11, 2003

3451

Figure 2. Structures of 10 hydroxycinnamic acid conjugates in eggplant fruit extracts that were identified or tentatively identified and represent four

groups classified as chlorogenic acid isomers (group 1), isochlorogenic acid isomers (group 2), hydroxycinnamic acid amide conjugates (group 3), and
acetylated chlorogenic acid isomers (group 5). LC peak numbers (P-1P-14) correspond to those indicated in the HPLC chromatogram in Figure 1.

species (15, 32). On the basis of its [M - 1]- ion (m/z 468)
and 1H NMR spectrum, the 20.6 min compound was identified
as N,N-dicaffeoylspermidine (C25H31N3O6 ) 469) (Figure 2).
1H NMR data were as follows (CD OD; ): 1.91-1.97 (6H,
3
m) [C2-H, C3-H, C6-H], 3.00-3.07 (8H, m) [C1-H, C4H, C5-H, C7-H], 6.35 (1H, d, J ) 16 Hz) [C8-H], 6.36 (1H,
d, J ) 16 Hz) [C8-H], 6.76 (2H, d, J ) 8 Hz) [C5-H, C5H], 6.88 (1H, dd, J ) 2, 8 Hz) [C6-H], ], 6.91 (1H, dd, J )
2, 8 Hz) [C6-H], 7.00 (1H, d, J ) 2 Hz) [C2-H], 7.01 (1H,
d, J ) 2 Hz) [C2-H], 7.40 (1H, d, J ) 8 Hz) [C7-H], 7.43
(1H, d, J ) 8 Hz) [C7-H]. The 10.1 min phenolic was
tentatively identified as N-caffeoylputrescine (C13H18N2O3 )
250) (Figure 2) on the basis of its [M - 1]- ion (m/z 249), the
close similarity of its UV spectrum to that of N,N-dicaffeoylspermidine, and partial 1H NMR data in the 6-8 ppm region
consistent with a caffeic acid moiety exhibiting chemical shifts
similar to those noted in N,N-dicaffeoylspermidine.
The two phenolics of group 4 had elution times 22.3 and
25.6 min (Figure 1; P-7 and P-11). Tentative identification as
caffeic acid conjugates (other than esters of quinic acid) was
based solely on the similarity of their UV spectra to those of
the mono- and di-O-caffeoylquinic acid isomers in groups 1
and 2 (Table 1). Because both of these compounds were
typically very minor constituents and each overlapped with other
phenolics, it was not possible to isolate them or obtain reliable
mass spectra.
The final group of phenolics, group 5, was composed of two
compounds with elution times of 24.7 and 26.1 min (Figure
1; P-10 and P-12). Their UV spectra differed only slightly from
one another and were similar to that of chlorogenic acid (Table
1). The ES--MS spectra of both included m/z 233 at 100%
relative intensity, [M - 1]- at m/z 395, and a sodium adduct
ion at m/z 417. Complete 1H NMR data for the 24.7 min
phenolic were as follows (CD3OD; ): 2.08-2.19 (3H, m)
[C2eq-H, C2eq-H, C6ax-H], 2.29 (1H, dd, J ) 4, 13 Hz)
[C6eq-H], 3.57 (3H, s) [OdC-O-Me], 3.92 (1H, dd, J ) 3,
8 Hz) [C4-H], 5.30 (1H, m) [C3-H], 5.37 (1H, m) [C5-H],
6.24 (1H, d, J ) 16 Hz) [C8-H], 6.76 (1H, d, J ) 8 Hz) [C5H], 6.95 (1H, dd, J ) 2, 8 Hz) [C6-H], 7.04 (1H, d, J ) 2
Hz) [C2-H], 7.54 (1H, d, J ) 8 Hz) [C7-H]. Signals from

the five protons on C2, C5, C6, C7, and C8 in the 6-8 ppm
region matched those of the caffeoyl moiety in chlorogenic acid
with very slight shifts; signals from the C3, C4, and C5 quinic
acid protons indicated that both the 3-OH and the 5-OH were
esterified (28); and the 3-proton singlet at 3.57 ppm indicated
a probable ester-linked methyl group (33). These data together
with the mass spectrometric data are consistent with the structure
3-O-acetyl-5-O-caffeoylquinic acid, C18H20O10 ) 396 (Figure
2). The principal ion at m/z 233 in the ES--MS spectra of both
the 24.7 and the 26.1 min phenolics can be accounted for by
the 3-O-acetylquinic acid moiety, C9H14O7 ) 234. Considering
the relative abundance and retention times of the 5-O- and 4-Ocaffeoylquinic acid isomers in group 1, it is likely (but not yet
determined) that the 26.1 min compound is 3-O-acetyl-4-Ocaffeoylquinic acid (Figure 2).
Analysis of the 14 hydroxycinnamic acid conjugates in
extracts from the fruit of S. melongena revealed considerable
diversity in composition and content among the seven commercial cultivars and among tissue segments within the fruit
(Table 2). Wide variation in hydroxycinnamic acid content
among cultivars has been noted for many other fruits (15, 34).
Significant differences between cultivars were evident for
composition and content of groups 1, 2, 3, 5, and total
hydroxycinnamic acid conjugates (F value 3.79*, 12.41***, 96,89***, 5.77**, and 6.05**, respectively, where *, **, and ***
indicate significance at P e 0.05, 0.01, and 0.001, respectively).
Levels of group 4 caffeic acid conjugates were not significantly
different between cultivars (F value 2.05). Black Magic, Elondo,
and Classic contained the highest levels of phenolics. Total
phenolic acid content was lowest in the cultivar Orient Express,
consistent with the perception by consumers that the Asian type
eggplants have a milder, less bitter flavor in comparison with
other market types.
As reported by Winter and Herrmann (9), chlorogenic acid
isomers in group 1 were the major class of hydroxycinnamic
acid conjugates, with their content varying 2.3-fold among the
cultivars evaluated. Group 1 accounted for 77.6-94.9% of the
total conjugates and comprised a greater percentage in cultivars
containing low levels of total phenolic acids compared with
those having high levels (Table 2). With the exception of

3452

J. Agric. Food Chem., Vol. 51, No. 11, 2003

Whitaker and Stommel

Table 2. Composition and Content of Hydroxycinnamic Acid Conjugates in Stem End, Midsection, and Blossom End Segments of Fruit Flesh from

S. melongena Cultivars
mol/100 g dry weight; (% of total)
cultivar
Black Magic

Elondo

Classic

Epic

Ghostbuster

Pirouette

Orient Express

slicea

group 1b

group 2

group 3

group 4

group 5

total

1
2
3
x
1
2
3
x
1
2
3
x
1
2
3
x
1
2
3
x
1
2
3
x
1
2
3
x

1875 (77.4)
3298 (79.2)
2868 (75.9)
2680 (77.6)
1692 (76.9)
3298 (83.7)
2715 (79.9)
2569 (80.8)
1494 (92.3)
3249 (91.3)
3402 (89.4)
2715 (90.7)
1107 (83.9)
2128 (81.5)
1562 (77.9)
1599 (80.8)
1009 (92.7)
2343 (88.6)
1783 (85.4)
1712 (88.2)
1314 (96.7)
2030 (94.5)
2011 (94.3)
1785 (94.9)
638 (95.1)
1448 (95.0)
1506 (93.9)
1197 (94.5)

5.4 (0.2)
9.9 (0.2)
43.8 (1.2)
19.7 (0.6)
1.0 (0.04)
3.2 (0.1)
35.2 (1.0)
13.1 (0.4)
2.3 (0.1)
21.3 (0.6)
68.9 (1.8)
30.8 (1.0)
0.6 (0.04)
8.0 (0.3)
25.9 (1.3)
11.5 (0.6)
1.1 (0.1)
8.3 (0.3)
27.5 (1.3)
12.3 (0.6)
1.5 (0.1)
1.2 (0.1)
6.4 (0.3)
3.0 (0.2)
1.5 (0.2)
0.2 (0.01)
1.2 (0.1)
1.0 (0.1)

512.0 (21.1)
782.8 (18.8)
815.8 (21.6)
703.6 (20.4)
484.8 (22.0)
562.7 (14.3)
595.0 (17.5)
547.5 (17.2)
94.2 (5.8)
214.3 (6.0)
273.1 (7.2)
193.9 (6.5)
186.6 (14.1)
411.7 (15.8)
369.7 (18.4)
322.6 (16.3)
60.2 (5.5)
169.4 (6.4)
194.9 (9.3)
141.5 (7.3)
28.3 (2.1)
95.2 (4.4)
89.0 (4.2)
70.8 (3.8)
20.3 (3.0)
46.6 (3.1)
60.6 (3.8)
42.5 (3.4)

26.5 (1.1)
49.3 (1.2)
43.9 (1.2)
39.9 (1.2)
21.8 (1.0)
37.8 (1.0)
33.4 (1.0)
31.0 (1.0)
27.0 (1.7)
35.9 (1.0)
37.4 (1.0)
33.4 (1.1)
23.5 (1.8)
39.8 (1.5)
29.2 (1.5)
30.8 (1.6)
16.5 (1.5)
24.3 (0.9)
20.5 (1.0)
20.5 (1.1)
13.4 (1.0)
17.6 (0.8)
19.5 (0.9)
16.9 (0.9)
10.8 (1.6)
28.9 (1.9)
34.6 (2.2)
24.8 (2.0)

1.9 (0.1)
24.8 (0.6)
8.1 (0.2)
11.6 (0.3)
1.4 (0.1)
39.0 (1.0)
17.7 (0.5)
19.4 (0.6)
1.4 (0.1)
39.8 (1.1)
25.7 (0.7)
22.3 (0.7)
1.6 (0.1)
24.0 (0.9)
17.8 (0.9)
14.5 (0.7)
1.0 (0.1)
99.4 (3.8)
60.6 (2.9)
53.7 (2.8)
1.3 (0.1)
4.3 (0.2)
7.3 (0.3)
4.3 (0.2)
0.8 (0.1)
1.1 (0.1)
1.4 (0.1)
1.1 (0.1)

2421
4165
3779
3455
2201
3941
3397
3180
1618
3560
3807
2995
1319
2612
2005
1979
1088
2644
2087
1940
1359
2148
2133
1880
671
1524
1604
1267

Slices 13 were taken from the stem end, midsection, and blossom end, respectively, of a longitudinal tissue segment cut from the middle of each fruit; x is the mean
value for all three slices. b Group 1, chlorogenic acid isomers; group 2, isochlorogenic acid isomers; group 3, amide conjugates; group 4, unknown caffeic acid conjugates;
and group 5, acetylated chlorogenic acid isomers.

Classic, the content of group 2 isochlorogenic acid isomers (3,5di-O > 4,5-di-O-caffeoylquinic acid) and group 3 hydroxycinnamic acid amide conjugates decreased as levels of chlorogenic
isomers declined. Much less variation in content among cultivars
was noted for group 4 caffeic acid derivatives (2.5-fold) than
for conjugates in groups 2, 3, or 5 (30.8-, 16.6-, and 48.8-fold,
respectively). Second to chlorogenic acid isomers, group 3 amide
conjugates accounted for the bulk of the additional phenolic
acids (3.4-20.4%) in all cultivars. The levels of these compounds can be highly dynamic, and various types of biotic and
abiotic stress can increase their synthesis (35, 36). Although
these factors cannot be disregarded, the 16.6-fold range in amide
conjugate content appears to be largely a function of cultivar
and the level of total phenolic acids. The remaining three groups,
isochlorogenic acid isomers, group 4 caffeic acid conjugates,
and acetylated chlorogenic acid isomers, comprised only 0.12.8% of total hydroxycinnamic acid conjugates in fruit of the
seven cultivars.
Significant differences between fruit stem and blossom end
tissues and between fruit stem and midsection tissues were
observed among all five groups of hydroxycinnamic acid
conjugates and in total conjugates across cultivars (Tables 2
and 3). Only isochlorogenic acid isomers and acetylated
chlorogenic acid isomers differed significantly between the fruit
midsection and the blossom end; the former were most abundant
in the blossom end, whereas the latter were most concentrated
in the midsection. The relatively high total hydroxycinnamic
acid content in the fruit midsection and blossom end could be
associated with seed development or may simply reflect the
advanced physiological maturity of these tissues as compared
with the fruit stem end. Isochlorogenic acid isomers and
acetylated chlorogenic acid isomers exhibited the greatest

Table 3. Variation in Content of Total Hydroxycinnamic Acid

Conjugates in Stem End, Middle, and Blossom End Segments of

Eggplant Fruit (n ) 62 for Slices 13 in Each Group)

groupa
1
2
3
4
5
total

slice b

mean content
(mol/100 g
dry wt)

1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3

1304
2550
2268
1.9
7.5
30.2
200.8
329.8
346.7
20.0
33.6
31.4
1.3
33.7
20.0
1528
2954
2696

t valuec
1 and 2

t valuec
2 and 3

t valuec
1 and 3

7.59***
1.46NS
3.78***
2.79**

6.02***
0.32NS

4.93***
0.68NS
5.93***
2.15*
7.09***
1.11NS

6.22***
8.10***
3.14**
4.14***
5.69***
6.17***

a Group 1, chlorogenic acid isomers; group 2, isochlorogenic acid isomers; group

3, amide conjugates; group 4, unknown caffeic acid conjugates; and group 5,
acetylated chlorogenic acid isomers. b Slices 13 were from the stem end, middle,
and blossom end, respectively. c t Value for differences between means for slices
1 and 2, 2 and 3, and 1 and 3. NS, *, **, *** signify nonsignificant or significant at P
e 0.05, 0.01, and 0.001, respectively.

variation from stem end to midsection and blossom end (25.9and 15.9-fold increase, respectively), whereas group 1 chlorogenic acid isomers, group 3 amide conjugates, and group 4

Hydroxycinnamic Acid Conjugates in Eggplant Fruit

caffeic acid conjugates showed much smaller increases of 2.0-,
1.7-, and 1.7-fold, respectively. Possibly, enzymes responsible
for the additional esterification of 5-O- or 4-O-caffeoylquinic
acid to yield these diesters are all but absent in stem end tissue
or, alternatively, the chlorogenic acids may have to reach a
certain threshold level before diester synthesis becomes significant.
In conclusion, this study determined that the content of
potentially health beneficial hydroxycinnamic acid conjugates
is substantial in eggplant fruit from commercial cultivars,
ranging from about 0.5-1.5% on a dry weight basis. This is in
accord with the value of about 600-660 mg/kg fresh weight
reported by Winter and Herrmann (9) and puts eggplant on a
par with sweet cherry, kiwi, and several other fruits that are
among the highest in phenolic acid content (15, 34). It was also
found that the levels and distribution of the individual conjugates
vary considerably among fruit from different cultivars and in
different tissue zones within each fruit. The content and
distribution of hydroxycinnamic acid derivatives in fruit of
various eggplant cultivars are likely to be influenced by climate,
cultural practices, harvest maturity, and postharvest storage
conditions. Total polyphenols declined by about 20-50% in
eggplant fruit stored for 20 days at 5, 10, or 20 C (11) and
increased by about 30-45% in fruit of three market types
between 30 and 42 days after fruit set (42 days being considered
physiological maturity) (12). Although chlorogenic acid isomers
comprised the bulk of the hydroxycinnamic acid derivatives in
fruit of all cultivars and in all tissues, several less abundant
phenolics showed the highest degree of variability among
cultivars and fruit tissues. Two of these were tentatively
identified as 3-O-acetyl esters of 5-O- and 4-O-caffeoylquinic
acid. Confirmation of these structures will require correlation
spectroscopy and 13C NMR analyses, but if correct, this appears
to be the first report of these phenolics in any plant tissue.
Should they prove to be of particular pharmacological or
nutritional interest, it is noteworthy that they were recently found
to be abundant in the fruit of a wild eggplant species, Solanum
anguiVi (Stommel and Whitaker, unpublished). Plant polyphenolics are the least well-known but quantitatively the most
important phytochemicals (34). Because various phytochemicals
have different functions and are distributed differently within
the body, combinations of compounds are likely to afford greater
health protective effects than any individual phytonutrient (4,
6). Our evaluation of eggplant hydroxycinnamic acid conjugates
provides data that will be useful in future health-related
epidemiological studies and human trials to ascertain the
influence of this class of phenolics on human health and in
development of new cultivars with optimal hydroxycinnamic
acid composition and content.
ACKNOWLEDGMENT

We thank Dr. Clifford Rice, USDA, ARS, Environmental

Quality Laboratory, for acquisition of the ES--MS mass spectra
and Dr. Walter F. Schmidt, USDA, ARS, Environmental Quality
Laboratory, for acquisition of the 1H NMR spectra.
LITERATURE CITED
(1) Huang, M. T.; Ferraro, T. Phenolic compounds in food and
cancer prevention. In Phenolic Compounds in Foods and their
Effects on Health: Antioxidant Cancer PreVention; Huang, M.
T., Ho, C. T., Lee, C. Y., Eds.; ACS Symposium Series 5078;
American Chemical Society: Washington, DC, 1992; pp 220238.

J. Agric. Food Chem., Vol. 51, No. 11, 2003

3453

(2) Ames, B. N.; Shigenaga, M. K.; Hagen, T. M. Oxidants,

antioxidants, and the degenerative diseases of aging. Proc. Natl.
Acad. Sci. U.S.A. 1993, 90, 7915-7922.
(3) Kinsella, J. E.; Frankel, E.; German, B.; Kanner, J. Possible
mechanisms for the protective role of antioxidants in wine and
plant foods. Food Technol. 1993, 47, 85-89.
(4) Steinmetz, K. A.; Potter, J. D. Vegetables, fruit, and cancer
prevention. J. Am. Diet. Assoc. 1996, 96, 1027-1039.
(5) Block, G.; Patterson, B.; Subar, A. Fruits, vegetables and cancer
prevention: a review of the epidemiological evidence. Nutr.
Cancer 1992, 18, 1-29.
(6) Rhodes, M. J. C. Bioactive components of food. Biochem. Soc.
Trans. 1996, 24, 771-835.
(7) Cao, G.; Sofic, E.; Prior, R. L. Antioxidant capacity of tea and
common vegetables. J. Agric. Food Chem. 1996, 44, 34263431.
(8) Sudheesh, S.; Presannakumar, G.; Vijayakumar, S.; Vijayalakshmi, N. R. Hypolipidemic effect of flavonoids from Solanum
melongena. Plant Foods Hum. Nutr. 1997, 51, 321-330.
(9) Winter, M.; Herrmann, K. Esters and glucosides of hydroxycinnamic acids in vegetables. J. Agric. Food Chem. 1986, 34,
616-620.
(10) Flick, G. J.; Burnette, F. S.; Aung, L. H.; Ory, R. L.; St. Angelo,
A. J. Chemical composition and biochemical properties of
mirlitons (Sechium edule) and purple, green, and white eggplants
(Solanum melongena). J. Agric. Food Chem. 1978, 26, 10001005.
(11) Esteban, R. M.; Molla, E.; Villarroya, M. B.; Lopez-Andreu, F.
J. Changes in the chemical composition of eggplant fruits during
storage. Sci. Hortic. 1989, 41, 19-25.
(12) Esteban, R. M.; Molla, E. M.; Robredo, L. M.; Lopez-Andreu,
F. J. Changes in the chemical composition of eggplant fruits
during development and ripening. J. Agric. Food Chem. 1992,
40, 998-1000.
(13) Croteau, R.; Kutchan, T. M.; Lewis, N. G. Natural products
(Secondary metabolites). In Biochemistry and Molecular Biology
of Plants; Buchanan, B. B., Gruissem, W., Jones, R. L., Eds.;
American Society of Plant Physiologists: Rockville, MD, 2000;
pp 1250-1318.
(14) Robards, K.; Prenzler, P. D.; Tucker, G.; Swatsitang, P.; Glover,
W. Phenolic compounds and their role in oxidative processes in
fruits. Food Chem. 1999, 66, 401-436.
(15) Machiex, J.-J.; Fleuriet, A.; Billot, J. Fruit Phenolics; CRC
Press: Boca Raton, FL, 1990; pp 1-104.
(16) Mlgaard, P.; Ravn, H. Evolutionary aspects of caffeoyl ester
distribution in dicotyledons. Phytochemistry 1988, 27, 24112421.
(17) Sawa, T.; Nakao, M.; Akaike, T.; Ono, K.; Maeda, H. Alkylperoxyl radical-scavenging activity of various flavonoids and other
phenolic compounds: Implications for the anti-tumor-promoter
effect of vegetables. J. Agric. Food Chem. 1999, 47, 397-402.
(18) Nakatani, N.; Kayano, S.; Kikuzaki, H.; Sumino, K.; Katagiri,
K.; Mitani, T. Identification, quantitative determination, and
antioxidative activities of chlorogenic acid isomers in prune
(Prunus domestica L.). J. Agric. Food Chem. 2000, 48, 55125516.
(19) Lekse, J. M.; Xia, L.; Morrow, J. D.; May, J. M. Plant catechols
prevent lipid peroxidation in human plasma erythrocytes. Mol.
Cell. Biochem. 2001, 226, 89-95.
(20) Nardini, M.; DAquino, M.; Tomassi, G.; Gentili, V.; Di Felice,
M.; Scaccini, C. Inhibition of human low-density lipoprotein
oxidation by caffeic acid and other hydroxycinnamic acid
derivatives. Free Radical Biol. Med. 1995, 19, 541-552.
(21) Cheminat, A.; Zawatzsky, R.; Becker, H.; Brouillard, R. Caffeoyl
conjugates from Echinacea species: Structures and biological
activity. Phytochemistry 1988, 27, 2787-2794.
(22) Robinson, W. E.; Cordeiro, M.; Abdel-Malek, S.; Jia, Q.; Chow,
S. A.; Reinecke, M. G.; Mitchell, W. M. Dicaffeoylquinic acid
inhibitors of human immunodeficiency virus integrase: inhibition
of the core catalytic domain of human immunodeficiency virus
integrase. Mol. Pharmacol. 1996, 50, 846-855.

3454

J. Agric. Food Chem., Vol. 51, No. 11, 2003

(23) Yoshimoto, M.; Okuno, S.; Islam, M. S.; Yamakawa. Polyphenolic content and antimutagenicity of sweetpotato leaves in
relation to commercial vegetables. In Program of the XXVI
International Horticultural Congress: Toronto, Canada, 2002;
p 264.
(24) Hongpattarakere, T.; Johnson, E. A. Natural antimicrobial
components isolated from yerba mate (Ilex paraguariensis). Food
Research Institute Annual Report; University of Wisconsin:
Madison, WI, 1999; p 39.
(25) University of Maryland. Commercial vegetable production
recommendations. UniV. Maryland Coop. Ext. SerV. Bull. 2000,
236.
(26) Buta, J. G.; Spaulding, D. W. Endogenous levels of phenolics
in tomato fruit during growth and maturation. Plant Growth
Regul. 1997, 16, 43-46.
(27) Schrader, K.; Kiehne, A.; Engelhardt, U. H.; Maier, H. G.
Determination of chlorogenic acids with lactones in roasted
coffee. J. Agric. Food Chem. 1996, 71, 392-398.
(28) Morishita, H.; Iwahashi, H.; Osaka, N.; Kido, R. Chromatographic separation and identification of naturally occurring
chlorogenic acids by 1H nuclear magnetic resonance spectroscopy
and mass spectrometry. J. Chromatogr. 1984, 315, 253-260.
(29) Whitaker, B. D.; Schmidt, W. F.; Kirk, M. C.; Barnes, S. Novel
fatty acid esters of p-coumaryl alcohol in epicuticular wax of
apple fruit. J. Agric. Food Chem. 2001, 49, 3787-3792.
(30) Amimoto, K.; Fukui, H. Quantitative differences in biologically
active components among hydroponically cultured lettuce and
endive cultivars. Acta Hortic. 1996, 440, 338-343.

Whitaker and Stommel

(31) Islam, M. S.; Yoshimoto, M.; Yahara, S.; Okuno, S.; Ishiguro,
K.; Yamakawa, O. Identification and characterization of foliar
polyphenolic composition in sweetpotato (Ipomoea batatas L.)
genotypes. J. Agric. Food Chem. 2002, 50, 3718-3722.
(32) Leubner-Metzger, G.; Amrhein, N. The distribution of hydroxycinnamoylputrescines in different organs of Solanum tuberosum
and other Solanaceous species. Phytochemistry 1993, 32, 551556.
(33) Merfort, I. Caffeoylquinic acids from flowers of Arnica montana
and Arnica chamissonis. Phytochemistry 1992, 31, 2111-2113.
(34) King, A.; Young, G. Characteristics and occurrence of phenolic
phytochemicals. J. Am. Diet. Assoc. 1999, 99, 213-218.
(35) Keinanen, M.; Oldham, N. J.; Baldwin, I. T. Rapid HPLC
screening of jasmonate-induced increases in tobacco alkaloids,
phenolics, and diterpene glycosides in Nicotiana attenuata. J.
Agric. Food Chem. 2001, 49, 3553-3558.
(36) Keller, H.; Hohlfeld, H.; Wray, V.; Hahlbrock, K.; Scheel, D.;
Strack, D. Changes in the accumulation of soluble and cell wallbound phenolics in elicitor-treated cell suspension cultures and
fungus-infected leaves of Solanum tuberosum. Phytochemistry
1996, 42, 389-396.
Received for review December 30, 2002. Revised manuscript received
March 31, 2003. Accepted April 1, 2003.

JF026250B